| { |
| "best_global_step": 600, |
| "best_metric": 0.27892550826072693, |
| "best_model_checkpoint": "/workspace/anchor/qwen-compassion-s1/train/checkpoint-600", |
| "epoch": 1.0, |
| "eval_steps": 50, |
| "global_step": 614, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.008143322475570033, |
| "grad_norm": 46.8648567199707, |
| "learning_rate": 2.5806451612903226e-05, |
| "loss": 3.4841, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.016286644951140065, |
| "grad_norm": 31.06366539001465, |
| "learning_rate": 5.8064516129032266e-05, |
| "loss": 3.5903, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.024429967426710098, |
| "grad_norm": 25.122819900512695, |
| "learning_rate": 9.032258064516129e-05, |
| "loss": 2.4155, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.03257328990228013, |
| "grad_norm": 3.6073243618011475, |
| "learning_rate": 0.00012258064516129034, |
| "loss": 1.5066, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.04071661237785016, |
| "grad_norm": 2.691239356994629, |
| "learning_rate": 0.00015483870967741937, |
| "loss": 1.0947, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.048859934853420196, |
| "grad_norm": 7.537977695465088, |
| "learning_rate": 0.0001870967741935484, |
| "loss": 0.8421, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.057003257328990226, |
| "grad_norm": 2.6401917934417725, |
| "learning_rate": 0.00019998693330688282, |
| "loss": 0.8682, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.06514657980456026, |
| "grad_norm": 4.235571384429932, |
| "learning_rate": 0.00019990709365929677, |
| "loss": 0.6972, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.0732899022801303, |
| "grad_norm": 2.4884767532348633, |
| "learning_rate": 0.00019975473152335726, |
| "loss": 0.6625, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.08143322475570032, |
| "grad_norm": 1.7398655414581299, |
| "learning_rate": 0.0001995299574986306, |
| "loss": 0.6196, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.08143322475570032, |
| "eval_loss": 0.44776251912117004, |
| "eval_runtime": 252.2446, |
| "eval_samples_per_second": 0.999, |
| "eval_steps_per_second": 0.5, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.08957654723127036, |
| "grad_norm": 1.4873219728469849, |
| "learning_rate": 0.00019923293474842174, |
| "loss": 0.6319, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.09771986970684039, |
| "grad_norm": 1.2676302194595337, |
| "learning_rate": 0.00019886387888133413, |
| "loss": 0.5349, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.10586319218241043, |
| "grad_norm": 1.0993133783340454, |
| "learning_rate": 0.00019842305779475968, |
| "loss": 0.4464, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.11400651465798045, |
| "grad_norm": 1.4379057884216309, |
| "learning_rate": 0.0001979107914804122, |
| "loss": 0.605, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.12214983713355049, |
| "grad_norm": 1.0843253135681152, |
| "learning_rate": 0.00019732745179204552, |
| "loss": 0.4638, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.13029315960912052, |
| "grad_norm": 1.7907661199569702, |
| "learning_rate": 0.00019667346217552527, |
| "loss": 0.4557, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.13843648208469056, |
| "grad_norm": 1.3890773057937622, |
| "learning_rate": 0.00019594929736144976, |
| "loss": 0.6569, |
| "step": 85 |
| }, |
| { |
| "epoch": 0.1465798045602606, |
| "grad_norm": 1.1585749387741089, |
| "learning_rate": 0.00019515548302054335, |
| "loss": 0.5085, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.15472312703583063, |
| "grad_norm": 3.5996532440185547, |
| "learning_rate": 0.0001942925953820725, |
| "loss": 0.4884, |
| "step": 95 |
| }, |
| { |
| "epoch": 0.16286644951140064, |
| "grad_norm": 1.0630170106887817, |
| "learning_rate": 0.00019336126081556134, |
| "loss": 0.5944, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16286644951140064, |
| "eval_loss": 0.3463878035545349, |
| "eval_runtime": 126.9731, |
| "eval_samples_per_second": 1.985, |
| "eval_steps_per_second": 0.992, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.17100977198697068, |
| "grad_norm": 1.1126106977462769, |
| "learning_rate": 0.00019236215537611046, |
| "loss": 0.4973, |
| "step": 105 |
| }, |
| { |
| "epoch": 0.1791530944625407, |
| "grad_norm": 1.198370099067688, |
| "learning_rate": 0.00019129600431364897, |
| "loss": 0.5198, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.18729641693811075, |
| "grad_norm": 1.4560389518737793, |
| "learning_rate": 0.00019016358154647618, |
| "loss": 0.4901, |
| "step": 115 |
| }, |
| { |
| "epoch": 0.19543973941368079, |
| "grad_norm": 1.185957908630371, |
| "learning_rate": 0.00018896570909947475, |
| "loss": 0.4398, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.20358306188925082, |
| "grad_norm": 0.9986463785171509, |
| "learning_rate": 0.00018770325650740345, |
| "loss": 0.4413, |
| "step": 125 |
| }, |
| { |
| "epoch": 0.21172638436482086, |
| "grad_norm": 1.1194801330566406, |
| "learning_rate": 0.00018637714018370253, |
| "loss": 0.5124, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.21986970684039087, |
| "grad_norm": 0.6920457482337952, |
| "learning_rate": 0.00018498832275526988, |
| "loss": 0.4722, |
| "step": 135 |
| }, |
| { |
| "epoch": 0.2280130293159609, |
| "grad_norm": 0.7126299738883972, |
| "learning_rate": 0.00018353781236369064, |
| "loss": 0.4714, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.23615635179153094, |
| "grad_norm": 1.1382180452346802, |
| "learning_rate": 0.00018202666193342833, |
| "loss": 0.5135, |
| "step": 145 |
| }, |
| { |
| "epoch": 0.24429967426710097, |
| "grad_norm": 1.3005266189575195, |
| "learning_rate": 0.00018045596840750723, |
| "loss": 0.5737, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.24429967426710097, |
| "eval_loss": 0.32419368624687195, |
| "eval_runtime": 122.2815, |
| "eval_samples_per_second": 2.061, |
| "eval_steps_per_second": 1.03, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.252442996742671, |
| "grad_norm": 0.8857340216636658, |
| "learning_rate": 0.0001788268719512427, |
| "loss": 0.4964, |
| "step": 155 |
| }, |
| { |
| "epoch": 0.26058631921824105, |
| "grad_norm": 1.2428998947143555, |
| "learning_rate": 0.00017714055512459565, |
| "loss": 0.4645, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.2687296416938111, |
| "grad_norm": 0.8355961441993713, |
| "learning_rate": 0.0001753982420237533, |
| "loss": 0.522, |
| "step": 165 |
| }, |
| { |
| "epoch": 0.2768729641693811, |
| "grad_norm": 1.104817271232605, |
| "learning_rate": 0.00017360119739255852, |
| "loss": 0.3735, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.28501628664495116, |
| "grad_norm": 0.7112654447555542, |
| "learning_rate": 0.00017175072570443312, |
| "loss": 0.5226, |
| "step": 175 |
| }, |
| { |
| "epoch": 0.2931596091205212, |
| "grad_norm": 0.7229294776916504, |
| "learning_rate": 0.00016984817021546177, |
| "loss": 0.4086, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.30130293159609123, |
| "grad_norm": 1.362148404121399, |
| "learning_rate": 0.000167894911989323, |
| "loss": 0.464, |
| "step": 185 |
| }, |
| { |
| "epoch": 0.30944625407166126, |
| "grad_norm": 1.0945537090301514, |
| "learning_rate": 0.00016589236889477646, |
| "loss": 0.3912, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.31758957654723124, |
| "grad_norm": 1.0569356679916382, |
| "learning_rate": 0.00016384199457643262, |
| "loss": 0.4927, |
| "step": 195 |
| }, |
| { |
| "epoch": 0.3257328990228013, |
| "grad_norm": 0.9064801931381226, |
| "learning_rate": 0.00016174527739955342, |
| "loss": 0.6048, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.3257328990228013, |
| "eval_loss": 0.3135300874710083, |
| "eval_runtime": 122.53, |
| "eval_samples_per_second": 2.057, |
| "eval_steps_per_second": 1.028, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.3338762214983713, |
| "grad_norm": 0.6857438087463379, |
| "learning_rate": 0.00015960373936964892, |
| "loss": 0.4339, |
| "step": 205 |
| }, |
| { |
| "epoch": 0.34201954397394135, |
| "grad_norm": 1.045210838317871, |
| "learning_rate": 0.0001574189350276545, |
| "loss": 0.5674, |
| "step": 210 |
| }, |
| { |
| "epoch": 0.3501628664495114, |
| "grad_norm": 1.105928897857666, |
| "learning_rate": 0.00015519245032149083, |
| "loss": 0.5271, |
| "step": 215 |
| }, |
| { |
| "epoch": 0.3583061889250814, |
| "grad_norm": 0.8385962247848511, |
| "learning_rate": 0.0001529259014548253, |
| "loss": 0.456, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.36644951140065146, |
| "grad_norm": 1.028890609741211, |
| "learning_rate": 0.00015062093371387097, |
| "loss": 0.4643, |
| "step": 225 |
| }, |
| { |
| "epoch": 0.3745928338762215, |
| "grad_norm": 1.2492626905441284, |
| "learning_rate": 0.00014827922027307451, |
| "loss": 0.4623, |
| "step": 230 |
| }, |
| { |
| "epoch": 0.38273615635179153, |
| "grad_norm": 0.9312857389450073, |
| "learning_rate": 0.00014590246098055996, |
| "loss": 0.4718, |
| "step": 235 |
| }, |
| { |
| "epoch": 0.39087947882736157, |
| "grad_norm": 0.7798976898193359, |
| "learning_rate": 0.00014349238112421024, |
| "loss": 0.3516, |
| "step": 240 |
| }, |
| { |
| "epoch": 0.3990228013029316, |
| "grad_norm": 0.772210419178009, |
| "learning_rate": 0.000141050730179282, |
| "loss": 0.4227, |
| "step": 245 |
| }, |
| { |
| "epoch": 0.40716612377850164, |
| "grad_norm": 1.088253140449524, |
| "learning_rate": 0.0001385792805384625, |
| "loss": 0.5712, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.40716612377850164, |
| "eval_loss": 0.3056293725967407, |
| "eval_runtime": 122.3692, |
| "eval_samples_per_second": 2.059, |
| "eval_steps_per_second": 1.03, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.4153094462540717, |
| "grad_norm": 0.9491221904754639, |
| "learning_rate": 0.00013607982622529133, |
| "loss": 0.4208, |
| "step": 255 |
| }, |
| { |
| "epoch": 0.4234527687296417, |
| "grad_norm": 1.0985082387924194, |
| "learning_rate": 0.00013355418159187985, |
| "loss": 0.5076, |
| "step": 260 |
| }, |
| { |
| "epoch": 0.43159609120521175, |
| "grad_norm": 0.6458399891853333, |
| "learning_rate": 0.00013100418000187419, |
| "loss": 0.4969, |
| "step": 265 |
| }, |
| { |
| "epoch": 0.43973941368078173, |
| "grad_norm": 0.9466511607170105, |
| "learning_rate": 0.0001284316724996181, |
| "loss": 0.4663, |
| "step": 270 |
| }, |
| { |
| "epoch": 0.44788273615635177, |
| "grad_norm": 0.9920846223831177, |
| "learning_rate": 0.00012583852646648095, |
| "loss": 0.5848, |
| "step": 275 |
| }, |
| { |
| "epoch": 0.4560260586319218, |
| "grad_norm": 1.0359402894973755, |
| "learning_rate": 0.00012322662426532708, |
| "loss": 0.4887, |
| "step": 280 |
| }, |
| { |
| "epoch": 0.46416938110749184, |
| "grad_norm": 0.6389766335487366, |
| "learning_rate": 0.00012059786187410984, |
| "loss": 0.351, |
| "step": 285 |
| }, |
| { |
| "epoch": 0.4723127035830619, |
| "grad_norm": 0.646231472492218, |
| "learning_rate": 0.00011795414750958265, |
| "loss": 0.3059, |
| "step": 290 |
| }, |
| { |
| "epoch": 0.4804560260586319, |
| "grad_norm": 0.8749302625656128, |
| "learning_rate": 0.00011529740024212565, |
| "loss": 0.4014, |
| "step": 295 |
| }, |
| { |
| "epoch": 0.48859934853420195, |
| "grad_norm": 1.0541212558746338, |
| "learning_rate": 0.00011262954860269399, |
| "loss": 0.5533, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48859934853420195, |
| "eval_loss": 0.2993938624858856, |
| "eval_runtime": 121.4763, |
| "eval_samples_per_second": 2.074, |
| "eval_steps_per_second": 1.037, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.496742671009772, |
| "grad_norm": 1.0241320133209229, |
| "learning_rate": 0.0001099525291828986, |
| "loss": 0.4102, |
| "step": 305 |
| }, |
| { |
| "epoch": 0.504885993485342, |
| "grad_norm": 1.0281414985656738, |
| "learning_rate": 0.00010726828522923562, |
| "loss": 0.4406, |
| "step": 310 |
| }, |
| { |
| "epoch": 0.5130293159609121, |
| "grad_norm": 1.0531939268112183, |
| "learning_rate": 0.00010457876523248518, |
| "loss": 0.5412, |
| "step": 315 |
| }, |
| { |
| "epoch": 0.5211726384364821, |
| "grad_norm": 0.5204165577888489, |
| "learning_rate": 0.00010188592151330343, |
| "loss": 0.3922, |
| "step": 320 |
| }, |
| { |
| "epoch": 0.5293159609120521, |
| "grad_norm": 0.60438072681427, |
| "learning_rate": 9.919170880503415e-05, |
| "loss": 0.455, |
| "step": 325 |
| }, |
| { |
| "epoch": 0.5374592833876222, |
| "grad_norm": 0.9122986793518066, |
| "learning_rate": 9.649808283476941e-05, |
| "loss": 0.2411, |
| "step": 330 |
| }, |
| { |
| "epoch": 0.5456026058631922, |
| "grad_norm": 0.6265031695365906, |
| "learning_rate": 9.38069989036886e-05, |
| "loss": 0.4733, |
| "step": 335 |
| }, |
| { |
| "epoch": 0.5537459283387622, |
| "grad_norm": 0.6672234535217285, |
| "learning_rate": 9.112041046770653e-05, |
| "loss": 0.252, |
| "step": 340 |
| }, |
| { |
| "epoch": 0.5618892508143323, |
| "grad_norm": 0.9693001508712769, |
| "learning_rate": 8.844026771946147e-05, |
| "loss": 0.4349, |
| "step": 345 |
| }, |
| { |
| "epoch": 0.5700325732899023, |
| "grad_norm": 0.9943844676017761, |
| "learning_rate": 8.57685161726715e-05, |
| "loss": 0.5285, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.5700325732899023, |
| "eval_loss": 0.2918320894241333, |
| "eval_runtime": 123.0632, |
| "eval_samples_per_second": 2.048, |
| "eval_steps_per_second": 1.024, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.5781758957654723, |
| "grad_norm": 0.76749587059021, |
| "learning_rate": 8.310709524988743e-05, |
| "loss": 0.4239, |
| "step": 355 |
| }, |
| { |
| "epoch": 0.5863192182410424, |
| "grad_norm": 0.9113866686820984, |
| "learning_rate": 8.045793687466757e-05, |
| "loss": 0.3481, |
| "step": 360 |
| }, |
| { |
| "epoch": 0.5944625407166124, |
| "grad_norm": 0.9299578666687012, |
| "learning_rate": 7.782296406919557e-05, |
| "loss": 0.5066, |
| "step": 365 |
| }, |
| { |
| "epoch": 0.6026058631921825, |
| "grad_norm": 1.1744991540908813, |
| "learning_rate": 7.520408955836007e-05, |
| "loss": 0.343, |
| "step": 370 |
| }, |
| { |
| "epoch": 0.6107491856677525, |
| "grad_norm": 0.5753459334373474, |
| "learning_rate": 7.260321438130913e-05, |
| "loss": 0.3693, |
| "step": 375 |
| }, |
| { |
| "epoch": 0.6188925081433225, |
| "grad_norm": 0.7324634194374084, |
| "learning_rate": 7.002222651148714e-05, |
| "loss": 0.2791, |
| "step": 380 |
| }, |
| { |
| "epoch": 0.6270358306188925, |
| "grad_norm": 1.022972822189331, |
| "learning_rate": 6.746299948615631e-05, |
| "loss": 0.5786, |
| "step": 385 |
| }, |
| { |
| "epoch": 0.6351791530944625, |
| "grad_norm": 0.7870802283287048, |
| "learning_rate": 6.492739104639727e-05, |
| "loss": 0.3502, |
| "step": 390 |
| }, |
| { |
| "epoch": 0.6433224755700325, |
| "grad_norm": 0.6990848183631897, |
| "learning_rate": 6.24172417885762e-05, |
| "loss": 0.4283, |
| "step": 395 |
| }, |
| { |
| "epoch": 0.6514657980456026, |
| "grad_norm": 0.9602301120758057, |
| "learning_rate": 5.9934373828257106e-05, |
| "loss": 0.3524, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.6514657980456026, |
| "eval_loss": 0.28751713037490845, |
| "eval_runtime": 122.7559, |
| "eval_samples_per_second": 2.053, |
| "eval_steps_per_second": 1.026, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.6596091205211726, |
| "grad_norm": 0.5980207920074463, |
| "learning_rate": 5.7480589477529545e-05, |
| "loss": 0.373, |
| "step": 405 |
| }, |
| { |
| "epoch": 0.6677524429967426, |
| "grad_norm": 0.811944305896759, |
| "learning_rate": 5.505766993671162e-05, |
| "loss": 0.4417, |
| "step": 410 |
| }, |
| { |
| "epoch": 0.6758957654723127, |
| "grad_norm": 1.0306131839752197, |
| "learning_rate": 5.266737400137765e-05, |
| "loss": 0.3625, |
| "step": 415 |
| }, |
| { |
| "epoch": 0.6840390879478827, |
| "grad_norm": 1.0983723402023315, |
| "learning_rate": 5.031143678565005e-05, |
| "loss": 0.3334, |
| "step": 420 |
| }, |
| { |
| "epoch": 0.6921824104234527, |
| "grad_norm": 0.5463206171989441, |
| "learning_rate": 4.799156846268095e-05, |
| "loss": 0.3236, |
| "step": 425 |
| }, |
| { |
| "epoch": 0.7003257328990228, |
| "grad_norm": 1.0159002542495728, |
| "learning_rate": 4.57094530232389e-05, |
| "loss": 0.4708, |
| "step": 430 |
| }, |
| { |
| "epoch": 0.7084690553745928, |
| "grad_norm": 0.8902086615562439, |
| "learning_rate": 4.346674705330117e-05, |
| "loss": 0.4642, |
| "step": 435 |
| }, |
| { |
| "epoch": 0.7166123778501629, |
| "grad_norm": 0.9829715490341187, |
| "learning_rate": 4.1265078531538916e-05, |
| "loss": 0.3721, |
| "step": 440 |
| }, |
| { |
| "epoch": 0.7247557003257329, |
| "grad_norm": 0.9822947978973389, |
| "learning_rate": 3.9106045647569e-05, |
| "loss": 0.4208, |
| "step": 445 |
| }, |
| { |
| "epoch": 0.7328990228013029, |
| "grad_norm": 1.1726772785186768, |
| "learning_rate": 3.69912156418289e-05, |
| "loss": 0.5286, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.7328990228013029, |
| "eval_loss": 0.283965528011322, |
| "eval_runtime": 122.8474, |
| "eval_samples_per_second": 2.051, |
| "eval_steps_per_second": 1.026, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.741042345276873, |
| "grad_norm": 1.0264843702316284, |
| "learning_rate": 3.492212366791831e-05, |
| "loss": 0.4188, |
| "step": 455 |
| }, |
| { |
| "epoch": 0.749185667752443, |
| "grad_norm": 0.9076156616210938, |
| "learning_rate": 3.290027167823204e-05, |
| "loss": 0.4011, |
| "step": 460 |
| }, |
| { |
| "epoch": 0.757328990228013, |
| "grad_norm": 0.9076895713806152, |
| "learning_rate": 3.092712733369387e-05, |
| "loss": 0.3638, |
| "step": 465 |
| }, |
| { |
| "epoch": 0.7654723127035831, |
| "grad_norm": 0.6383405327796936, |
| "learning_rate": 2.9004122938382617e-05, |
| "loss": 0.3438, |
| "step": 470 |
| }, |
| { |
| "epoch": 0.7736156351791531, |
| "grad_norm": 0.8013618588447571, |
| "learning_rate": 2.7132654399823444e-05, |
| "loss": 0.4277, |
| "step": 475 |
| }, |
| { |
| "epoch": 0.7817589576547231, |
| "grad_norm": 0.8031596541404724, |
| "learning_rate": 2.531408021569982e-05, |
| "loss": 0.4388, |
| "step": 480 |
| }, |
| { |
| "epoch": 0.7899022801302932, |
| "grad_norm": 0.8627935647964478, |
| "learning_rate": 2.3549720487720738e-05, |
| "loss": 0.5149, |
| "step": 485 |
| }, |
| { |
| "epoch": 0.7980456026058632, |
| "grad_norm": 0.8749674558639526, |
| "learning_rate": 2.184085596336011e-05, |
| "loss": 0.4311, |
| "step": 490 |
| }, |
| { |
| "epoch": 0.8061889250814332, |
| "grad_norm": 0.9885497689247131, |
| "learning_rate": 2.0188727106162874e-05, |
| "loss": 0.455, |
| "step": 495 |
| }, |
| { |
| "epoch": 0.8143322475570033, |
| "grad_norm": 0.7557974457740784, |
| "learning_rate": 1.8594533195293427e-05, |
| "loss": 0.4357, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8143322475570033, |
| "eval_loss": 0.2808151841163635, |
| "eval_runtime": 123.1626, |
| "eval_samples_per_second": 2.046, |
| "eval_steps_per_second": 1.023, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8224755700325733, |
| "grad_norm": 0.8533985614776611, |
| "learning_rate": 1.7059431454979824e-05, |
| "loss": 0.4475, |
| "step": 505 |
| }, |
| { |
| "epoch": 0.8306188925081434, |
| "grad_norm": 4.26806116104126, |
| "learning_rate": 1.5584536214485457e-05, |
| "loss": 0.3914, |
| "step": 510 |
| }, |
| { |
| "epoch": 0.8387622149837134, |
| "grad_norm": 1.0256925821304321, |
| "learning_rate": 1.4170918099218166e-05, |
| "loss": 0.4249, |
| "step": 515 |
| }, |
| { |
| "epoch": 0.8469055374592834, |
| "grad_norm": 0.7930788397789001, |
| "learning_rate": 1.2819603253564205e-05, |
| "loss": 0.317, |
| "step": 520 |
| }, |
| { |
| "epoch": 0.8550488599348535, |
| "grad_norm": 0.7796340584754944, |
| "learning_rate": 1.153157259601062e-05, |
| "loss": 0.3707, |
| "step": 525 |
| }, |
| { |
| "epoch": 0.8631921824104235, |
| "grad_norm": 0.9223588109016418, |
| "learning_rate": 1.030776110709718e-05, |
| "loss": 0.4599, |
| "step": 530 |
| }, |
| { |
| "epoch": 0.8713355048859935, |
| "grad_norm": 0.880706250667572, |
| "learning_rate": 9.149057150714801e-06, |
| "loss": 0.4563, |
| "step": 535 |
| }, |
| { |
| "epoch": 0.8794788273615635, |
| "grad_norm": 0.9193106889724731, |
| "learning_rate": 8.056301829242784e-06, |
| "loss": 0.3882, |
| "step": 540 |
| }, |
| { |
| "epoch": 0.8876221498371335, |
| "grad_norm": 0.5757290720939636, |
| "learning_rate": 7.030288372993066e-06, |
| "loss": 0.2638, |
| "step": 545 |
| }, |
| { |
| "epoch": 0.8957654723127035, |
| "grad_norm": 0.9016059637069702, |
| "learning_rate": 6.0717615644051206e-06, |
| "loss": 0.4252, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.8957654723127035, |
| "eval_loss": 0.2792342007160187, |
| "eval_runtime": 122.54, |
| "eval_samples_per_second": 2.056, |
| "eval_steps_per_second": 1.028, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.9039087947882736, |
| "grad_norm": 0.8001049160957336, |
| "learning_rate": 5.181417197408734e-06, |
| "loss": 0.4541, |
| "step": 555 |
| }, |
| { |
| "epoch": 0.9120521172638436, |
| "grad_norm": 0.7635800838470459, |
| "learning_rate": 4.359901572347758e-06, |
| "loss": 0.4125, |
| "step": 560 |
| }, |
| { |
| "epoch": 0.9201954397394136, |
| "grad_norm": 0.9592556357383728, |
| "learning_rate": 3.607811026831176e-06, |
| "loss": 0.4653, |
| "step": 565 |
| }, |
| { |
| "epoch": 0.9283387622149837, |
| "grad_norm": 0.9132296442985535, |
| "learning_rate": 2.9256915028519573e-06, |
| "loss": 0.3343, |
| "step": 570 |
| }, |
| { |
| "epoch": 0.9364820846905537, |
| "grad_norm": 0.8796995878219604, |
| "learning_rate": 2.3140381504882737e-06, |
| "loss": 0.4447, |
| "step": 575 |
| }, |
| { |
| "epoch": 0.9446254071661238, |
| "grad_norm": 0.7750846147537231, |
| "learning_rate": 1.7732949684743594e-06, |
| "loss": 0.3234, |
| "step": 580 |
| }, |
| { |
| "epoch": 0.9527687296416938, |
| "grad_norm": 1.1729882955551147, |
| "learning_rate": 1.303854481902067e-06, |
| "loss": 0.4184, |
| "step": 585 |
| }, |
| { |
| "epoch": 0.9609120521172638, |
| "grad_norm": 0.9538452625274658, |
| "learning_rate": 9.060574572873237e-07, |
| "loss": 0.4235, |
| "step": 590 |
| }, |
| { |
| "epoch": 0.9690553745928339, |
| "grad_norm": 0.9514711499214172, |
| "learning_rate": 5.801926552078563e-07, |
| "loss": 0.4121, |
| "step": 595 |
| }, |
| { |
| "epoch": 0.9771986970684039, |
| "grad_norm": 1.043609380722046, |
| "learning_rate": 3.264966206921294e-07, |
| "loss": 0.4091, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.9771986970684039, |
| "eval_loss": 0.27892550826072693, |
| "eval_runtime": 123.7565, |
| "eval_samples_per_second": 2.036, |
| "eval_steps_per_second": 1.018, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.9853420195439739, |
| "grad_norm": 1.0530821084976196, |
| "learning_rate": 1.451535115114866e-07, |
| "loss": 0.4419, |
| "step": 605 |
| }, |
| { |
| "epoch": 0.993485342019544, |
| "grad_norm": 0.7235980033874512, |
| "learning_rate": 3.629496450011649e-08, |
| "loss": 0.3433, |
| "step": 610 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 614, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 100, |
| "stateful_callbacks": { |
| "EarlyStoppingCallback": { |
| "args": { |
| "early_stopping_patience": 2, |
| "early_stopping_threshold": 0.001 |
| }, |
| "attributes": { |
| "early_stopping_patience_counter": 1 |
| } |
| }, |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.508719122176676e+18, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
